Miao Zhang

An Improved Genetic-Based Approach to Task Scheduling in Inter-cloud Environment

2015

With the development of cloud computing, the number of cloud computing service providers has arisen rapidly. The research of task scheduling in cloud computing environment nearly enters a mature stage. But when there is a …

Stratify Mobile App Reviews: E-LDA Model Based on Hot "Entity" Discovery

2016

Recent literatures have illustrated approaches that can automatically extract informative content from noisy mobile app reviews, however the key information such as feature requests, bug reports etc., retrieved by these methods are still mixed and …

Overcoming Multi-Model Forgetting in One-Shot NAS With Diversity Maximization

2020

One-Shot Neural Architecture Search (NAS) significantly improves the computational efficiency through weight sharing. However, this approach also introduces multi-model forgetting during the supernet training (architecture search phase), where the performance of previous architectures degrade when …

Key Exchange and Management Schemes for Automotive Control Units

2021

Today's vehicles are typically equipped with dozens of electronic control units that control one or more of the electrical systems or subsystems. Due to the rapid development towards autonomous driving, vehicle functionalities strongly depend on …

Differential Privacy Data Release Scheme Using Microaggregation With Conditional Feature Selection

2023 · IEEE Internet of Things Journal

Differential privacy (DP) has achieved great progress in addressing the user privacy preservation issues related to data analysis in the Internet of Things (IoT) services and applications. However, the existing DP models tend to overlook …

Text Segmentation Algorithm Focused on Corpus Mining for Oilfield Exploration and Development

2024

The professional knowledge of oilfield exploration and development is vast and complex, and manual organization is time-consuming. Inspired by natural language large models and sequential modeling, a long text segmentation algorithm focused on corpus mining …

OledFL: Unleashing the Potential of Decentralized Federated Learning via Opposite Lookahead Enhancement

2024 · arXiv (Cornell University)

Decentralized Federated Learning (DFL) surpasses Centralized Federated Learning (CFL) in terms of faster training, privacy preservation, and light communication, making it a promising alternative in the field of federated learning. However, DFL still exhibits significant …

FALCON: Resolving Visual Redundancy and Fragmentation in High-Resolution Multimodal Large Language Models via Visual Registers

2025

The incorporation of high-resolution visual input equips multimodal large language models (MLLMs) with enhanced visual perception capabilities for real-world tasks. However, most existing high-resolution MLLMs rely on a cropping-based approach to process images, which leads …

AgentDropout: Dynamic Agent Elimination for Token-Efficient and High-Performance LLM-Based Multi-Agent Collaboration

2025

Multi-agent systems (MAS) based on large language models (LLMs) have demonstrated significant potential in collaborative problemsolving.However, they still face substantial challenges of low communication efficiency and suboptimal task performance, making the careful design of the …

DTI with Minimal Data: Image Translation Based Distortion Correction and FA Map Generation for Clinical Efficiency

2025 · Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition

Motivation: Current DTI processing requires around 30 directions per shell to ensure data quality, which is difficult to obtain in certain vulnerable patient groups. Goal(s): To reduce number of DWI directions needed in clinical study. …

From Technological Capabilities to Emotional Illusions: a First Look at the Mechanisms Mediating the Phenomenon of AI Hallucinations

2025 · Creativity and Innovation

This paper focuses on the phenomenon of "hallucination" of artificial intelligence (AI) in the context of media communication, pointing out that the current AI application is undergoing a transition process from technical functionality to emotional …

JTreeformer: Graph-Transformer via Latent-Diffusion Model for Molecular Generation

2025 · arXiv (Cornell University)

The discovery of new molecules based on the original chemical molecule distributions is of great importance in medicine. The graph transformer, with its advantages of high performance and scalability compared to traditional graph networks, has …